Machine Learning: a new era for cardiovascular pregnancy physiology and cardio-obstetrics research.

The maternal cardiovascular system undergoes functional and structural adaptations during pregnancy and postpartum to support increased metabolic demands of offspring and placental growth, labor, and delivery, as well as recovery from childbirth. Pregnancy thus poses physiological stress upon the maternal cardiovascular system, and in the absence of an appropriate response it imparts potential risks for cardiovascular complications and adverse outcomes. The proportion of pregnancy-related maternal deaths from cardiovascular events has been steadily increasing, contributing to high rates of maternal mortality. Despite advances in cardiovascular physiology research, there is still no comprehensive understanding of maternal cardiovascular adaptations in healthy pregnancies, and with far less known about pregnancy with complications. Further, current tools for prognosis of cardiovascular complications during pregnancy are limited. Machine learning (ML) offers new and effective tools for investigating mechanisms involved in pregnancy-related cardiovascular complications as well as the development of potential therapies. The main goal of this review is to summarize existing research that uses ML to understand mechanisms of cardiovascular physiology during pregnancy and develop prediction models for clinical application in pregnant patients. We also provide an overview of ML fundamentals and a discussion about platforms that can be used to enhance understanding of cardiovascular adaptations to pregnancy. Finally, we address the interpretability and explainability of ML outcomes, consequences of model bias, and ethics of ML use.

Emergency department use in the postpartum period: a retrospective cohort study.

Background: Visits to the emergency room (ED) by women in the postpartum period may reflect gaps in postpartum care and disparities in access to obstetric and primary care services. This study aimed to characterize the patients who visited the ED in the first year after delivery, their reasons for coming to the ED, and the care they received. Methods: The electronic health record was reviewed for all patients who delivered at University of Iowa Health Care between 2009 and 2023 and visited the ED within 365 days after delivery. Data drawn directly from the EHR included patient demographics and medical history, pregnancy and delivery information, and newborn characteristics. The charts were then reviewed manually for information regarding ED visits including time from delivery, chief complaint, diagnosis, and disposition. Results: 555 pregnancies had ED visits within one year of delivery, with a total 814 ED visits across the study sample. 46.7% of ED visits occurred in the first 30 days following delivery, and 35% of ED visits for obstetric complaints occurred in the first 2 weeks after delivery. Black patients visited the ED more often (mean=1.84 visits, SD=1.30) than white (mean=1.34, SD=0.92, p<0.001) or Hispanic patients (mean=1.35, SD=0.67, p = 0.004). The most common categories of chief complaint were obstetric (34.6%) and gastrointestinal (18.8%), while the most common categories of diagnosis were obstetric (31.8%) and immune/infectious (28.1%). Conclusions: Visits to the ED are common in the year following delivery. Almost half of these visits occur in the first 30 days after birth. The plurality of postpartum ED visits are due to obstetric complaints, especially in the first few weeks. Black women are more likely to use the ED during this period, potentially due to disparities in healthcare access. These findings suggest that some of these ED visits may be preventable, and that there is room for improvement in post-delivery follow-up, communication between patients and the obstetrics team, and access to outpatient obstetric care.

A role for adverse childhood experiences and depression in preeclampsia.

Introduction: Adverse childhood experiences (ACEs) are a measure of childhood adversity and are associated with life-long morbidity. The impacts of ACEs on peripartum health including preeclampsia, a common and dangerous hypertensive disorder of pregnancy, remain unclear, however. Therefore, we aimed to determine ACE association with peripartum psychiatric health and prevalence of preeclampsia using a case-control design. Methods: Clinical data were aggregated and validated using a large, intergenerational knowledgebase developed at our institution. Depression symptoms were measured by standard clinical screeners: the Patient Health Questionnaire-9 (PHQ-9) and the Edinburgh Postnatal Depression Scale (EPDS). ACEs were assessed via survey. Scores were compared between participants with (N = 32) and without (N = 46) prior preeclampsia. Results: Participants with ACE scores ≥4 had significantly greater odds of preeclampsia than those with scores ≤ 3 (adjusted odds ratio = 6.71, 95% confidence interval:1.13-40.00; p = 0.037). Subsequent speculative analyses revealed that increased odds of preeclampsia may be driven by increased childhood abuse and neglect dimensions of the ACE score. PHQ-9 scores (3.73 vs. 1.86, p = 0.03), EPDS scores (6.38 vs. 3.71, p = 0.01), and the incidence of depression (37.5% vs. 23.9%, p = 0.05) were significantly higher in participants with a history of preeclampsia versus controls. Conclusions: Childhood sets the stage for life-long health. Our findings suggest that ACEs may be a risk factor for preeclampsia and depression, uniting the developmental origins of psychiatric and obstetric risk.

Predicting the onset of preeclampsia by longitudinal monitoring of metabolic changes throughout pregnancy with Raman spectroscopy.

Preeclampsia is a life-threatening pregnancy disorder. Current clinical assays cannot predict the onset of preeclampsia until the late 2nd trimester, which often leads to poor maternal and neonatal outcomes. Here we show that Raman spectroscopy combined with machine learning in pregnant patient plasma enables rapid, highly sensitive maternal metabolome screening that predicts preeclampsia as early as the 1st trimester with >82% accuracy. We identified 12, 15 and 17 statistically significant metabolites in the 1st, 2nd and 3rd trimesters, respectively. Metabolic pathway analysis shows multiple pathways corresponding to amino acids, fatty acids, retinol, and sugars are enriched in the preeclamptic cohort relative to a healthy pregnancy. Leveraging Pearson's correlation analysis, we show for the first time with Raman Spectroscopy that metabolites are associated with several clinical factors, including patients' body mass index, gestational age at delivery, history of preeclampsia, and severity of preeclampsia. We also show that protein quantification alone of proinflammatory cytokines and clinically relevant angiogenic markers are inadequate in identifying at-risk patients. Our findings demonstrate that Raman spectroscopy is a powerful tool that may complement current clinical assays in early diagnosis and in the prognosis of the severity of preeclampsia to ultimately enable comprehensive prenatal care for all patients.

Low indoleamine 2, 3 dioxygenase (IDO) activity is associated with psycho-obstetric risk.

OBJECTIVES: Preeclampsia and depression in pregnancy are among the most prevalent obstetric disorders with no known cures. While depression and preeclampsia each increase risk for the other, shared mechansisms are unclear. One possibility is low levels of Indoleamine 2,3 dioxygenase (IDO), which links immune dysregulation and oxidative arterial damage resulting in poor vascular function in both preeclampsia and depression. We hypothesized low circulating IDO activity levels in pregnancy would correspond to poor vascular function and depression symptoms. STUDY DESIGN: In this nested case-control study, clinical, demographic, and biologic data from a cohort of pregnant women recruited to longitudinal studies measuring noninvasive vascular function and circulating factors were analyzed. MAIN OUTCOME MEASURE: IDO activity across all three trimesters of pregnancy was measured using a colorimetric assay. Carotid-femoral pulse wave velocity (cfPWV), a measure of arterial stiffness, was also assessed throughout gestation by non-invasive applanation tonometry. Depression symptoms were assessed in pregnancy via the validated patient health questionnaire 9 (PHQ9). RESULTS: Participants with low second and third trimester IDO activity had significantly decreased cfPWV. This association remained statistically significant when controlled for confounders such as BMI and chronic hypertension in the third but not second trimester. While PHQ9 scores were not associated with cfPWV differences, IDO activity was lower in moderate and severely depressed relative to non-depressed pregnant individuals. CONCLUSION: These results implicate IDO in arterial stiffness and depression symptoms, suggesting that decreased IDO may be a central target for improved psycho-obstetric health.

Study Design and Protocol of the Multisite Pregnancy 24/7 Cohort Study.

Hypertensive disorders of pregnancy and other adverse pregnancy outcomes (APOs) are associated with an increased risk of future maternal cardiovascular disease. Physical activity during pregnancy reduces the risk of these APOs, yet few meet physical activity guidelines during pregnancy. Little is known about the role of sedentary behavior or sleep in APOs, a critical gap in knowledge given these behaviors comprise the majority of a 24-hour day. To address this knowledge gap, the Pregnancy 24/7 cohort study (2020-2025) uses 2 devices for 24-hour activity assessment in each trimester of pregnancy to examine associations of sedentary behavior, sleep, and the 24-hour activity cycle (composition of sedentary behavior, physical activity, and sleep) with hypertensive disorders and other APOs. Participants (n = 500) are recruited from the University of Iowa, University of Pittsburgh, and West Virginia University in early pregnancy and followed through delivery. The activPAL3 micro and Actiwatch Spectrum Plus are worn in each trimester for 7 days of 24-hour wear to assess the 24-hour activity cycle. APOs are abstracted from medical charts. This study will provide critical data to fuel future research examining how modifying the 24-hour activity cycle in pregnancy can improve maternal health.

Anti-angiogenic mechanisms and serotonergic dysfunction in the Rgs2 knockout model for the study of psycho-obstetric risk.

Psychiatric and obstetric diseases are growing threats to public health and share high rates of co-morbidity. G protein-coupled receptor signaling (e.g., vasopressin, serotonin) may be a convergent psycho-obstetric risk mechanism. Regulator of G Protein Signaling 2 (RGS2) mutations increase risk for both the gestational disease preeclampsia and for depression. We previously found preeclampsia-like, anti-angiogenic obstetric phenotypes with reduced placental Rgs2 expression in mice. Here, we extend this to test whether conserved cerebrovascular and serotonergic mechanisms are also associated with risk for neurobiological phenotypes in the Rgs2 KO mouse. Rgs2 KO exhibited anxiety-, depression-, and hedonic-like behaviors. Cortical vascular density and vessel length decreased in Rgs2 KO; cortical and white matter thickness and cell densities were unchanged. In Rgs2 KO, serotonergic gene expression was sex-specifically changed (e.g., cortical Htr2a, Maoa increased in females but all serotonin targets unchanged or decreased in males); redox-related expression increased in paraventricular nucleus and aorta; and angiogenic gene expression was changed in male but not female cortex. Whole-cell recordings from dorsal raphe serotonin neurons revealed altered 5-HT1A receptor-dependent inhibitory postsynaptic currents (5-HT1A-IPSCs) in female but not male KO neurons. Additionally, serotonin transporter blockade by the SSRI sertraline increased the amplitude and time-to-peak of 5-HT1A-IPSCs in KO neurons to a greater extent than in WT neurons in females only. These results demonstrate behavioral, cerebrovascular, and sertraline hypersensitivity phenotypes in Rgs2 KOs, some of which are sex-specific. Disruptions may be driven by vascular and cell stress mechanisms linking the shared pathogenesis of psychiatric and obstetric disease to reveal future targets.

First Trimester Prediction of Preterm Birth in Patient Plasma with Machine-Learning-Guided Raman Spectroscopy and Metabolomics.

Preterm birth (PTB) is the leading cause of infant deaths globally. Current clinical measures often fail to identify women who may deliver preterm. Therefore, accurate screening tools are imperative for early prediction of PTB. Here, we show that Raman spectroscopy is a promising tool for studying biological interfaces, and we examine differences in the maternal metabolome of the first trimester plasma of PTB patients and those that delivered at term (healthy). We identified fifteen statistically significant metabolites that are predictive of the onset of PTB. Mass spectrometry metabolomics validates the Raman findings identifying key metabolic pathways that are enriched in PTB. We also show that patient clinical information alone and protein quantification of standard inflammatory cytokines both fail to identify PTB patients. We show for the first time that synergistic integration of Raman and clinical data guided with machine learning results in an unprecedented 85.1% accuracy of risk stratification of PTB in the first trimester that is currently not possible clinically. Correlations between metabolites and clinical features highlight the body mass index and maternal age as contributors of metabolic rewiring. Our findings show that Raman spectral screening may complement current prenatal care for early prediction of PTB, and our approach can be translated to other patient-specific biological interfaces.

Inflammatory Biomarker Profiles in Very Preterm Infants within the Context of Preeclampsia, Chorioamnionitis, and Clinically Diagnosed Postnatal Infection.

Preterm delivery can be precipitated by preeclampsia or infection, and preterm infants are at heightened risk of postnatal infection. Little is known about the ontogeny of inflammatory biomarkers in extremely preterm infants. We hypothesized that suspected prenatal infection (clinical chorioamnionitis or spontaneous preterm labor) and clinically diagnosed postnatal infection would be associated with unique biomarker signatures, and those patterns would be influenced by the degree of prematurity. Venous blood was collected daily for the first week and weekly for up to 14 additional weeks from 142 neonates born at 22-32 weeks gestation. A custom array was utilized to measure monocyte chemoattractant protein-1 (MCP-1) and interleukin-6 (IL-6). C-reactive protein (CRP) levels were obtained from the electronic medical record. Independent of gestational age, MCP-1 was significantly increased (p < 0.001) in association with maternal preeclampsia, but MCP-1 was decreased (p < 0.01), and CRP was increased (p < 0.01) in the presence of chorioamnionitis with funisitis. IL-6 and CRP were both increased in infants diagnosed with postnatal infection, with peak levels observed on days 2 and 3, respectively. In conclusion, suspected prenatal and postnatal infections and non-infectious complications of pregnancy are associated with unique biomarker profiles, independent of gestational age, including over a 2-fold increase in MCP-1 among newborns of mothers with preeclampsia. Further, in those clinically diagnosed with a postnatal infection in the absence of antenatal infection concerns, IL-6 increases before CRP, emphasizing a potential role for expanded biomarker screening if antibiotics are initially avoided in infants delivered for maternal indications.

Maternal and fetal mitochondrial gene dysregulation in hypertensive disorders of pregnancy.

Mitochondrial dysfunction has been implicated in pregnancy-induced hypertension (PIH). The role of mitochondrial gene dysregulation in PIH, and consequences for maternal-fetal interactions, remain elusive. Here, we investigated mitochondrial gene expression and dysregulation in maternal and placental tissues from pregnancies with and without PIH; further, we measured circulating mitochondrial DNA (mtDNA) mutational load, an index of mtDNA integrity. Differential gene expression analysis followed by Time Course Gene Set Analysis (TcGSA) was conducted on publicly available high throughput sequencing transcriptomic data sets. Mutational load analysis was carried out on peripheral mononuclear blood cells from healthy pregnant individuals and individuals with preeclampsia. Thirty mitochondrial differentially expressed genes (mtDEGs) were detected in the maternal cell-free circulating transcriptome, whereas nine were detected in placental transcriptome from pregnancies with PIH. In PIH pregnancies, maternal mitochondrial dysregulation was associated with pathways involved in inflammation, cell death/survival, and placental development, whereas fetal mitochondrial dysregulation was associated with increased production of extracellular vesicles (EVs) at term. Mothers with preeclampsia did not exhibit a significantly different degree of mtDNA mutational load. Our findings support the involvement of maternal mitochondrial dysregulation in the pathophysiology of PIH and suggest that mitochondria may mediate maternal-fetal interactions during healthy pregnancy.NEW & NOTEWORTHY This study identifies aberrant maternal and fetal expression of mitochondrial genes in pregnancies with gestational hypertension and preeclampsia. Mitochondrial gene dysregulation may be a common etiological factor contributing to the development of de novo hypertension in pregnancy-associated hypertensive disorders.

Virome Analysis and Association of Positive Coxsackievirus B Serology during Pregnancy with Congenital Heart Disease.

BACKGROUND: We have previously shown coxsackievirus B (CVB) to be a potent inducer of congenital heart disease (CHD) in mice. The clinical relevance of these findings in humans and the roles of other viruses in the pathogenesis of CHD remain unknown. METHODS: We obtained plasma samples, collected at all trimesters, from 89 subjects (104 pregnancies), 73 healthy controls (88 pregnancies), and 16 with CHD-affected birth (16 pregnancies), from the Perinatal Family Tissue Bank (PFTB). We performed CVB IgG/IgM serological assays on plasma. We also used ViroCap sequencing and PCR to test for viral nucleic acid in plasma, circulating leukocytes from the buffy coat, and in the media of a co-culture system. RESULTS: CVB IgG/IgM results indicated that prior exposure was 7.8 times more common in the CHD group (95% CI, 1.14-54.24, adj. p-value = 0.036). However, the CVB viral genome was not detected in plasma, buffy coat, or co-culture supernatant by molecular assays, although other viruses were detected. CONCLUSION: Detection of viral nucleic acid in plasma was infrequent and specifically no CVB genome was detected. However, serology demonstrated that prior CVB exposure is higher in CHD-affected pregnancies. Further studies are warranted to understand the magnitude of the contribution of the maternal blood virome to the pathogenesis of CHD.

Invasive or More Direct Measurements Can Provide an Objective Early-Stopping Ceiling for Training Deep Neural Networks on Non-invasive or Less-Direct Biomedical Data.

Early stopping is an extremely common tool to minimize overfitting, which would otherwise be a cause of poor generalization of the model to novel data. However, early stopping is a heuristic that, while effective, primarily relies on ad hoc parameters and metrics. Optimizing when to stop remains a challenge. In this paper, we suggest that for some biomedical applications, a natural dichotomy of invasive/non-invasive measurements, or more generally proximal vs distal measurements of a biological system can be exploited to provide objective advice on early stopping. We discuss the conditions where invasive measurements of a biological process should provide better predictions than non-invasive measurements, or at best offer parity. Hence, if data from an invasive measurement are available locally, or from the literature, that information can be leveraged to know with high certainty whether a model of non-invasive data is overfitted. We present paired invasive/non-invasive cardiac and coronary artery measurements from two mouse strains, one of which spontaneously develops type 2 diabetes, posed as a classification problem. Examination of the various stopping rules shows that generalization is reduced with more training epochs and commonly applied stopping rules give widely different generalization error estimates. The use of an empirically derived training ceiling is demonstrated to be helpful as added information to leverage early stopping in order to reduce overfitting.

Circulating microparticle proteins predict pregnancies complicated by placenta accreta spectrum.

Placenta accreta spectrum (PAS) is characterized by abnormal attachment of the placenta to the uterus, and attempts at placental delivery can lead to catastrophic maternal hemorrhage and death. Multidisciplinary delivery planning can significantly improve outcomes; however, current diagnostics are lacking as approximately half of pregnancies with PAS are undiagnosed prior to delivery. This is a nested case-control study of 35 cases and 70 controls with the primary objective of identifying circulating microparticle (CMP) protein panels that identify pregnancies complicated by PAS. Size exclusion chromatography and liquid chromatography with tandem mass spectrometry were used for CMP protein isolation and identification, respectively. A two-step iterative workflow was used to establish putative panels. Using plasma sampled at a median of 26 weeks' gestation, five CMP proteins distinguished PAS from controls with a mean area under the curve (AUC) of 0.83. For a separate sample taken at a median of 35 weeks' gestation, the mean AUC was 0.78. In the second trimester, canonical pathway analyses demonstrate over-representation of processes related to iron homeostasis and erythropoietin signaling. In the third trimester, these analyses revealed abnormal immune function. CMP proteins classify PAS well prior to delivery and have potential to significantly reduce maternal morbidity and mortality.

Selective Serotonin Reuptake Inhibitor Use in Pregnancy and Protective Mechanisms in Preeclampsia.

Depression and preeclampsia share risk factors and are bi-directionally associated with increased risk for each other. Despite epidemiological evidence linking selective serotonin reuptake inhibitors (SSRIs) in pregnancy to preeclampsia, serotonin (5-HT) and vasopressin (AVP) secretion mechanisms suggest that SSRIs may attenuate preeclampsia risk. However, there is a need to clarify the relationship between SSRIs and preeclampsia in humans to determine therapeutic potential. This retrospective cohort study included clinical data from 9558 SSRI-untreated and 9046 SSRI-treated pregnancies. In a subcohort of 233 pregnancies, early pregnancy (< 20 weeks) maternal plasma copeptin, an inert and stable AVP prosegment secreted 1:1 with AVP, was measured by enzyme-linked immunosorbent assay. Diagnoses and depression symptoms (Patient Health Questionnaire-9 [PHQ-9]) were identified via medical records review. Descriptive, univariate, and multivariate regression analyses were conducted (α = 0.05). SSRI use was associated with decreased preeclampsia after controlling for clinical confounders (depression severity, chronic hypertension, diabetes, body mass index, age) (OR = 0.9 [0.7-1.0], p = 0.05). Moderate-to-severe depression symptoms were associated with significantly higher copeptin secretion than mild-to-no depression symptoms (240 ± 29 vs. 142 ± 10 ng/mL, p < 0.001). SSRIs significantly attenuated first trimester plasma copeptin (78 ± 22 users vs. 240 ± 29 ng/ml non-users, p < 0.001). In preeclampsia, SSRI treatment was associated with significantly lower copeptin levels (657 ± 164 vs. 175 ± 134 ng/mL, p = 0.04). Interaction between SSRI treatment and preeclampsia was also significant (p = 0.04). SSRIs may modulate preeclampsia risk and mechanisms, although further studies are needed to investigate the relationships between 5-HT and AVP in depression and preeclampsia.

Effects of maternal hypertension on cord blood Arginine vasopressin receptor expression.

Arginine vasopressin (AVP) signaling is altered in preeclampsia and physiologic stress. AVP is implicated in fluid homeostasis and cardiovascular (CV) function, which is disrupted in some progeny from preeclamptic pregnancies. However, whether altered fetal AVP signaling occurs in preeclampsia is unknown. Here, we measured CV-related transcripts (e.g., AVP receptors) in cord blood via quantitative PCR. Chronic hypertension decreased AVPR1b, AVPR2, OXTR, LNPEP, and CUL5. AVPR1a, AVPR1b, and AVPR2 were decreased while OXTR was increased in preeclamptic cord blood. In sum, we found prenatal exposure to hypertension in pregnancy alters fetal AVP signaling and may thereby prime offspring CV disease risk.

Need for Improved Collection and Harmonization of Rural Maternal Healthcare Data.

Representation in data sets is critical to improving healthcare for the largest possible number of people. Unfortunately, pregnancy is a very understudied period of time. Further, the gap in available data is wide between pregnancies in urban areas versus rural areas. There are many limitations in the current data that is available. Herein, we review these limitations and strengths of available data sources. In addition, we propose a new mechanism to enhance the granularity, depth, and speed with which data is made available regarding rural pregnancy.

A Case of Persistent Human Pegivirus Infection in Two Separate Pregnancies of a Woman.

Human pegivirus (HPgV) is best known for persistent, presumably non-pathogenic, infection and a propensity to co-infect with human immunodeficiency virus or hepatitis C virus. However, unique attributes, such as the increased risk of malignancy or immune modulation, have been recently recognized for HPgV. We have identified a unique case of a woman with high levels HPgV infection in two pregnancies, which occurred 4 years apart and without evidence of human immunodeficiency virus or hepatitis C virus infection. The second pregnancy was complicated by congenital heart disease. A high level of HPgV infection was detected in the maternal blood from different trimesters by RT-PCR and identified as HPgV type 1 genotype 2 in both pregnancies. In the second pregnancy, the decidua and intervillous tissue of the placenta were positive for HPgV by PCR but not the chorion or cord blood (from both pregnancies), suggesting no vertical transmission despite high levels of viremia. The HPgV genome sequence was remarkably conserved over the 4 years. Using VirScan, sera antibodies for HPgV were detected in the first trimester of both pregnancies. We observed the same anti-HPgV antibodies against the non-structural NS5 protein in both pregnancies, suggesting a similar non-E2 protein humoral immune response over time. To the best of our knowledge, this is the first report of persistent HPgV infection involving placental tissues with no clear indication of vertical transmission. Our results reveal a more elaborate viral-host interaction than previously reported, expand our knowledge about tropism, and opens avenues for exploring the replication sites of this virus.

Selective serotonin reuptake inhibitors and preeclampsia: A quality assessment and meta-analysis.

Serotonin modulates vascular, immune, and neurophysiology and is dysregulated in preeclampsia. Despite biological plausibility that selective serotonin reuptake inhibitors (SSRIs) prevent preeclampsia pathophysiology, observational studies have indicated increased risk and providers may be hesitant. The objective of this meta-analysis and quality assessment was to evaluate the evidence linking SSRI use in pregnancy to preeclampsia/gestational hypertension. PubMed was searched through June 5, 2020 manually and using combinations of terms: "preeclampsia", "serotonin", and "SSRI". This review followed MOOSE guidelines. Inclusion criteria were: 1) Observational cohort or population study, 2) exposure defined as SSRI use during pregnancy, 3) cases defined as preeclampsia or gestational hypertension, and 4) human participants. Studies were selected that addressed the hypothesis that gestational SSRI use modulates preeclampsia and/or gestational hypertension risk. Review Manager Web was used to synthesize study findings. Articles were read and scored (Newcastle-Ottawa Quality Assessment Scale) for quality by two independent reviewers. Publication bias was assessed using a funnel plot and the Egger test. Of 179 screened studies, nine were included. The pooled risk ratio (random effects model) was 1.43 (95 % CI: 1.15-1.78, P < 0.001; range 0.96-4.86). Two studies were rated as moderate quality (both with total score of 6); others were high quality. Heterogeneity was high (I2 = 88 %) and funnel asymmetry was significant (p < 0.00001). Despite evidence for increased preeclampsia risk with SSRIs, shared risk factors and other variables are poorly controlled. Depression treatment should not be withheld due to perceived gestational hypertension risk. Mechanistic evidence for serotonin modulation in preeclampsia demonstrates a need for future research.

Implementation of a Maternal Child Knowledgebase.

To advance the application of clinical data to address maternal health we developed and implemented a Maternal Child Knowledgebase (MCK). The MCK integrates data from every pregnancy that received care at the University of Iowa Hospitals & Clinics (UIHC) and links information from the pregnancy episode to the delivery episode and between the mother and child. This knowledgebase contains integrated information regarding diagnoses, medications, mother and child vitals, hospital admissions, depression screenings, laboratory value results, and procedure information. It also collates information from the electronic health record (EPIC), the Social Security Death Index, and the Medication Administration Record into one knowledgebase. To enhance usability, we designed a custom viewer with several pre-designed queries and reports that eliminates the need for users to be proficient in SQL coding. The recent implementation of the MCK has supported multiple projects and reduced the number of Obstetrics-related data queries to the Biomedical Informatics group.